Available as of Camel 2.9
This component allows interaction with cloud provider key-value engines (blobstores) and compute services. The component uses jclouds which is
a library that provides abstractions for blobstores and compute services.
ComputeService simplifies the task of managing machines in the cloud. For example, you can use ComputeService to start 5 machines and install your software on them.
BlobStore simplifies dealing with key-value providers such as Amazon S3. For example, BlobStore can give you a simple Map view of a container.
The camel jclouds component allows you to use both abstractions, as it specifes two types of endpoint the JcloudsBlobStoreEndpoint and the JcloudsComputeEndpoint. You can have both producers and consumers on a blobstore endpoint but you can only have producers on compute endpoints.
Maven users will need to add the following dependency to their pom.xml
for this component:
<dependency> <groupId>org.apache.camel</groupId> <artifactId>camel-jclouds</artifactId> <version>x.x.x</version> <!-- use the same version as your Camel core version --> </dependency> |
The camel jclouds component will make use of multiple jclouds blobstores and compute services as long as they are passed to the component during initialization. The component accepts a list blobstores and compute services. Here is how it can be configured.
<bean id="jclouds" class="org.apache.camel.component.jclouds.JcloudsComponent"> <property name="computeServices"> <list> <ref bean="computeService"/> </list> </property> <property name="blobStores"> <list> <ref bean="blobStore"/> </list> </property> </bean> <!-- Creating a blobstore from spring / blueprint xml --> <bean id="blobStoreContextFactory" class="org.jclouds.blobstore.BlobStoreContextFactory"/> <bean id="blobStoreContext" factory-bean="blobStoreContextFactory" factory-method="createContext"> <constructor-arg name="provider" value="PROVIDER_NAME"/> <constructor-arg name="identity" value="IDENTITY"/> <constructor-arg name="credential" value="CREDENTIAL"/> </bean> <bean id="blobStore" factory-bean="blobStoreContext" factory-method="getBlobStore"/> <!-- Creating a compute service from spring / blueprint xml --> <bean id="computeServiceContextFactory" class="org.jclouds.compute.ComputeServiceContextFactory"/> <bean id="computeServiceContext" factory-bean="computeServiceContextFactory" factory-method="createContext"> <constructor-arg name="provider" value="PROVIDER_NAME"/> <constructor-arg name="identity" value="IDENTITY"/> <constructor-arg name="credential" value="CREDENTIAL"/> </bean> <bean id="computeService" factory-bean="computeServiceContext" factory-method="getComputeService"/> |
As you can see the component is capable of handling multiple blobstores and compute services. The actual implementation that will be used by each endpoint is specified by passing the provider inside the URI.
jclouds:blobstore:[provider id][?options] jclouds:compute:[provider id][?options] |
The provider id is the name of the cloud provider that provides the target service (e.g. aws-s3 or aws_ec2).
You can append query options to the URI in the following format, ?option=value&option=value&...
|
You can have as many of these options as you like.
jclouds:blobstore:aws-s3?operation=CamelJcloudsGet&container=mycontainer&blobName=someblob |
For producer endpoint you can override all of the above URI options by passing the appropriate headers to the message.
|
This example will show you how you can store any message inside a blob using the jclouds component.
from("direct:start") .to("jclouds:blobstore:aws-s3" + "?operation=PUT" + "&container=mycontainer" + "&blobName=myblob"); |
In the above example you can override any of the URI parameters with headers on the message.
Here is how the above example would look like using xml to define our route.
<route> <from uri="direct:start"/> <to uri="jclouds:blobstore:aws-s3?operation=PUT&container=mycontainer&blobName=myblob"/> </route> |
This example will show you how you can read the contnet of a blob using the jclouds component.
from("direct:start") .to("jclouds:blobstore:aws-s3" + "?operation=GET" + "&container=mycontainer" + "&blobName=myblob"); |
In the above example you can override any of the URI parameters with headers on the message.
Here is how the above example would look like using xml to define our route.
<route> <from uri="direct:start"/> <to uri="jclouds:blobstore:aws-s3?operation=PUT&container=mycontainer&blobName=myblob"/> </route> |
This example will consume all blob that are under the specified container. The generated exchange will contain the payload of the blob as body.
from("jclouds:blobstore:aws-s3" + "?container=mycontainer") .to("direct:next"); |
You can achieve the same goal by using xml, as you can see below.
<route> <from uri="jclouds:blobstore:aws-s3?operation=GET&container=mycontainer&blobName=myblob"/> <to uri="direct:next"/> </route> |
|
The combination of parameters for use with the compute service depend on the operation.
jclouds:compute:aws-ec2?operation=CamelJcloudsCreateNode&imageId=AMI_XXXXX&locationId=eu-west-1&group=mygroup |
Below are some examples that demonstrate the use of jclouds compute producer in java dsl and spring/blueprint xml.
from("jclouds:compute:aws-ec2" + "&operation=CamelJCloudsListImages") .to("direct:next"); |
This will create a message that will contain the list of images inside its body. You can also do the same using xml.
<route> <from uri="jclouds:compute:aws-ec2?operation=CamelJCloudsListImages"/> <to uri="direct:next"/> </route> |
from("direct:start"). to("jclouds:compute:aws-ec2" + "?operation=CamelJcloudsCreateNode" + "&imageId=AMI_XXXXX" + "&locationId=XXXXX" + "&group=myGroup"); |
This will create a new node on the cloud provider. The out message in this case will be a set of metadata that contains information about the newly created node (e.g. the ip, hostname etc). Here is the same using spring xml.
<route> <from uri="direct:start"/> <to uri="jclouds:compute:aws-ec2?operation=CamelJcloudsCreateNode&imageId=AMI_XXXXX&locationId=XXXXX&group=myGroup"/> </route> |
from("direct:start"). to("jclouds:compute:aws-ec2" + "?operation=CamelJcloudsRunScript" + "?nodeId=10" + "&user=ubuntu"); |
The sample above will retrieve the body of the in message, which is expected to contain the shell script to be executed. Once the script is retrieved, it will be sent to the node for execution under the specified user (in order case ubuntu). The target node is specified using its nodeId. The nodeId can be retrieved either upon the creation of the node, it will be part of the resulting metadata or by a executing a LIST_NODES operation.
Note This will require that the compute service that will be passed to the component, to be initialized with the appropriate jclouds ssh capable module (e.g. jsch or sshj).
Here is the same using spring xml.
<route> <from uri="direct:start"/> <to uri="jclouds:compute:aws-ec2?operation=CamelJcloudsListNodes&?nodeId=10&user=ubuntu"/> </route> |
If you want to find out more about jclouds here is list of interesting resources
Jclouds Blobstore wiki
Jclouds Compute wiki